A game guide generation method, device and equipment and storage medium

Abstract

Embodiments of the present application disclose a game guide generation method and device, equipment and a storage medium. The game guide generation method provided by the embodiments of the present application comprises: receiving user demand information and a style selection instruction, determining a guide style template according to the style selection instruction, wherein the user demand information comprises one or more of a game name, a game version, a game character, a guide direction and a target audience; constructing a knowledge graph based on a set knowledge base according to the user demand information to obtain a key prompt word; generating an executable guide generation instruction according to the key prompt word and the guide style template, and executing the guide generation instruction to generate a game guide of a corresponding style. The scheme can solve the technical problem of low game guide generation efficiency, and can automatically generate a game guide of a corresponding style based on user demand information and a style selection instruction, thereby significantly improving the game guide generation efficiency.

Description

Technical Field

[0001] This application relates to the field of game technology, and in particular to a method, apparatus, device and storage medium for generating game guides. Background Technology

[0002] With the booming development of the video game industry, the mechanics of large-scale open-world games, highly complex MMORPGs (massively multiplayer online role-playing games), and strategy games are becoming increasingly complex, significantly increasing players' reliance on game guides. Game guides, as the core information carrier for players to understand game rules, master character skills, and optimize tactical strategies, not only help beginners get started quickly but also provide experienced players with insights into cutting-edge gameplay in each version iteration, directly impacting players' gaming experience and competitive level.

[0003] Currently, game guide generation relies heavily on manual writing, but this method has significant limitations. Game guides need to integrate multi-dimensional information such as terminology definitions, skill values, combat strategies, and version updates. Professional guide authors need to invest a significant amount of time in understanding the complex in-game mechanics, resulting in a long production cycle. For example, guide authors need to manually sift through official game documentation, community discussions, and practical data. For frequently updated games, collecting data for version preview guides alone can take several days. Therefore, the existing method of manually writing game guides is relatively inefficient and cannot meet the rapidly evolving technological demands of the game industry. Summary of the Invention

[0004] This application provides a method, apparatus, device, and storage medium for generating game guides, which can solve the technical problem of low efficiency in generating game guides. Based on user needs information and style selection instructions, game guides of the corresponding style can be automatically generated, significantly improving the efficiency of game guide generation.

[0005] In a first aspect, embodiments of this application provide a method for generating game guides, including:

[0006] Receive user request information and style selection instructions, and determine the strategy style template according to the style selection instructions. The user request information includes one or more of the following: game name, game version, game character, strategy direction, and target audience.

[0007] Based on user needs and a set knowledge base, a knowledge graph is constructed to obtain key prompts.

[0008] Generate executable strategy generation commands based on key prompts and strategy style templates, and execute the strategy generation commands to generate game strategies in the corresponding style.

[0009] Furthermore, before constructing the knowledge graph based on the set knowledge base according to user needs information, it also includes:

[0010] Receive game reference materials, which include one or more of the following: game standard definitions, game encyclopedia data, cost guides for this game, guides for finished general games, and restriction rules;

[0011] The game reference materials are parsed to obtain metadata, and the metadata is annotated to generate the corresponding knowledge base.

[0012] Furthermore, based on user needs information and the established knowledge base, a knowledge graph is constructed to obtain key prompts, including:

[0013] Determine the search index based on user needs information;

[0014] Based on the set knowledge base, the search index is used to retrieve metadata and obtain the target triplet data;

[0015] A knowledge graph is constructed based on the target triple data, and key prompt words are extracted from the constructed knowledge graph.

[0016] Furthermore, executing the strategy generation command generates game strategies in the corresponding style, including:

[0017] Determine the text generation sub-instructions, image retrieval sub-instructions, and text-image integration sub-instructions within the strategy generation command;

[0018] Execute the text generation sub-command to generate game strategy text in the corresponding style;

[0019] Execute the image retrieval sub-instruction to retrieve the corresponding game screenshot from the set knowledge base based on the game strategy text, and obtain the target image;

[0020] Execute the image and text integration sub-command to merge the game strategy text and the target image to obtain the first target game strategy.

[0021] Furthermore, executing the strategy generation command generates game strategies in the corresponding style, including:

[0022] Identify the text generation sub-instructions, video acquisition sub-instructions, video compositing sub-instructions, and video-text fusion sub-instructions within the strategy generation command;

[0023] Execute the text generation sub-command to generate game strategy text in the corresponding style;

[0024] Execute the video acquisition sub-instruction to retrieve the corresponding target game video from the set knowledge base according to the user's requirements, and edit the target game video according to the key prompts to obtain the target video clip;

[0025] Execute the video compositing sub-instruction to dynamically composite the target video clips based on the game strategy text to obtain the initial strategy video;

[0026] Execute the video-text fusion sub-instruction to embed the game strategy text into the initial strategy video, resulting in the second target game strategy.

[0027] Furthermore, after executing the strategy generation command to generate a game strategy guide of the corresponding style, it also includes:

[0028] The interactive interface displays game guides along with their corresponding style and professionalism indices.

[0029] Receive adjustment instructions based on style index and professionalism index, and obtain target style index and target professionalism index according to the adjustment instructions;

[0030] Based on the target style index and target professionalism index, a new game guide is generated using user needs information and style selection instructions.

[0031] Furthermore, receiving user request information includes:

[0032] Receive user input including voice messages, text descriptions, or images;

[0033] The received voice information is parsed and processed to obtain the corresponding user demand information;

[0034] Alternatively, semantic analysis can be performed on the received text description to obtain the corresponding user demand information;

[0035] Alternatively, the received image can be intelligently recognized to obtain the corresponding user requirement information.

[0036] In a second aspect, embodiments of this application provide a game strategy generation apparatus, comprising:

[0037] The information receiving module is used to receive user requirement information and style selection instructions;

[0038] The template determination module is used to determine the strategy style template based on the style selection instruction. User requirement information includes one or more of the following: game name, game version, game character, strategy direction, and target audience.

[0039] The prompt word determination module is used to construct a knowledge graph based on a set knowledge base according to user needs information, and obtain key prompt words;

[0040] The strategy generation module is used to generate executable strategy generation commands based on key hints and strategy style templates, and then execute the strategy generation commands to generate game strategies in the corresponding style.

[0041] In a third aspect, embodiments of this application provide a game strategy generation device, comprising:

[0042] Memory and one or more processors;

[0043] Memory, used to store one or more programs;

[0044] When one or more programs are executed by one or more processors, the one or more processors implement the game strategy generation method as described in the first aspect.

[0045] In a fourth aspect, embodiments of this application provide a storage medium for storing computer-executable instructions, which, when executed by a computer processor, are used to perform the game strategy generation method as described in the first aspect.

[0046] This application embodiment determines a strategy style template based on a received style selection instruction during game strategy guide generation. It then constructs a knowledge graph based on a set knowledge base using received user request information to obtain key prompts. Finally, it generates an executable strategy guide generation instruction based on the key prompts and the strategy style template, and executes the instruction to generate a game strategy guide of the corresponding style. By employing this technique, game strategy guides of the corresponding style can be automatically generated based on user request information and style selection instructions. This avoids the technical problem of low efficiency in game strategy guide generation caused by manual writing, improves the automation and intelligence of game strategy guide generation, and significantly enhances the efficiency of game strategy guide generation.

[0047] The beneficial effects of the game strategy generation device, game strategy generation equipment, and storage medium provided above can be referenced in relation to the beneficial effects of the game strategy generation method. Attached Figure Description

[0048] Figure 1 This is a flowchart of a game strategy generation method provided in an embodiment of this application;

[0049] Figure 2 This is a flowchart illustrating another game strategy generation method provided in an embodiment of this application;

[0050] Figure 3 This is a first schematic diagram of a game strategy generation interactive interface provided in an embodiment of this application;

[0051] Figure 4 This is a flowchart illustrating another game strategy generation method provided in this application embodiment;

[0052] Figure 5 This is a flowchart illustrating another game strategy generation method provided in an embodiment of this application;

[0053] Figure 6This is a flowchart illustrating another game strategy generation method provided in this application embodiment;

[0054] Figure 7 This is a flowchart illustrating another game strategy generation method provided in an embodiment of this application;

[0055] Figure 8 This is a second schematic diagram of a game strategy generation interactive interface provided in an embodiment of this application;

[0056] Figure 9 This is a schematic diagram of the structure of a game strategy generation device provided in an embodiment of this application;

[0057] Figure 10 This is a schematic diagram of the structure of a game strategy generation device provided in an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, specific embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining this application and not for limiting it. It should also be noted that, for ease of description, only the parts relevant to this application are shown in the drawings, not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but may also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0059] Currently, game guide generation relies heavily on manual writing, but this method has significant limitations. Game guides need to integrate multi-dimensional information such as terminology definitions, skill values, combat strategies, and version updates. Professional guide authors need to invest a significant amount of time in understanding the complex in-game mechanics, resulting in a long production cycle. For example, guide authors need to manually sift through official game documentation, community discussions, and practical data. For frequently updated games, collecting data for version preview guides alone can take several days. Therefore, the existing method of manually writing game guides is relatively inefficient and cannot meet the rapidly evolving technological demands of the game industry.

[0060] Based on this, the present application provides a game guide generation method, apparatus, device, and storage medium. The aim of this application is to determine a guide style template based on a received style selection instruction during game guide generation, construct a knowledge graph based on a set knowledge base using received user requirement information to obtain key prompts, generate an executable guide generation instruction based on the key prompts and the guide style template, and execute the guide generation instruction to generate a game guide of the corresponding style. By employing the above technical means, compared to manually writing game guides, this embodiment can automatically generate game guides of the corresponding style using user requirement information and style selection instructions, improving the automation and intelligence of game guide generation and significantly increasing the efficiency of game guide generation.

[0061] Figure 1 A flowchart of a game guide generation method provided in this application embodiment is given. The game guide generation method provided in this embodiment can be executed by a game guide generation device, which can be implemented by software and / or hardware. The game guide generation device can be composed of two or more physical entities, or it can be composed of a single physical entity. Generally, the game guide generation device can be a computer device.

[0062] The following description uses a computer device as the primary example to illustrate the game strategy generation method. (Refer to...) Figure 1 The specific methods for generating this game guide include:

[0063] S11. Receive user requirement information and style selection instructions, and determine the strategy style template according to the style selection instructions. The user requirement information includes one or more of the following: game name, game version, game character, strategy direction, and target audience.

[0064] Users can input their needs and style selection instructions through a pre-defined interactive interface. The computer device receives these instructions, which include one or more of the following: game name, game version, game character, strategy direction, and target audience. The system can preset styles, such as "humorous," "serious," and "academic." Users can select from these preset styles through the interface, generating a style selection instruction based on their choice. The target style is then determined according to this instruction. Based on the mapping between the pre-defined styles and strategy style templates, the corresponding strategy style template is determined. For example, keywords can be extracted from the user's input text description. The most semantically similar target style can be selected from the preset styles based on the extracted keywords. A style selection instruction is generated based on the determined target style, and the corresponding strategy style template is determined based on the mapping between the pre-defined styles and strategy style templates. For example, if the user inputs a text description of "lighthearted and funny guide", the keywords extracted from the text description are "lighthearted" and "funny". After calculation, it is determined that the keyword with the closest semantic similarity to the preset style is "humorous style". Then, a style selection instruction for "humorous style" is generated, and the corresponding "humorous style" guide style template is matched according to the style selection instruction.

[0065] In one embodiment, user requirements can be input via voice, text, or images. Style selection instructions can be selected using preset controls or input via corresponding voice, text, or images. When there is ambiguity in the received user requirements and style selection instructions, follow-up questions can be used to complete the information, making the corresponding user requirements and style selection instructions clearer.

[0066] In one embodiment, user needs information includes one or more of the following: game name, game version, game character, strategy direction, and target audience. Strategy direction includes beginner's guides, advanced techniques, and specific level breakthroughs, while the target audience includes novice players, experienced players, and casual players. In addition, user needs information may also include game platform information, specific game element focus, desired strategy depth, and preferred presentation style. Different game platforms have different operating methods and hardware performance compatibility, therefore, the corresponding game strategies will also differ. Specific game element focus refers to the varying levels of attention players pay to different elements in the game. Some players focus on character development and want guides that detail character upgrades, skill bonuses, and equipment combinations; some players enjoy exploring the game map and need information on map resource distribution and hidden location locations; still others are interested in the game's story and desire guides that outline the plot and interpret the story's background and foreshadowing. Therefore, the corresponding game strategy can be determined based on the specific game element focus. The desired depth of strategy guides refers to the significant differences in users' needs for such depth. New players may only require basic gameplay introductions and walkthroughs, while experienced players seek in-depth guides covering advanced challenges, deep analysis of game mechanics, and extreme value calculations. Therefore, different game guides can be tailored to players' desired depth of strategy guides. Preferred presentation format refers to the presentation style of game guides. Some prefer pure text descriptions for their conciseness and clarity; others favor a combination of text and images to aid understanding of game scenarios and operational steps; still others find video guides more intuitive, clearly showing the actual gameplay process. Therefore, different formats of game guides can be determined based on players' preferred presentation formats.

[0067] As described above, by using style selection commands and corresponding strategy style templates, game strategy guides can be customized in style to meet the preferences of different players, thereby increasing the personalization of game strategy guides and improving the user experience. Furthermore, by using style selection commands to match the style of game strategy guides, the time spent manually writing guides is reduced, thus improving the efficiency of game strategy guide generation and further enhancing the user experience.

[0068] In one embodiment, users can also dynamically add custom-style strategy guide templates. Users can upload multiple sample texts with new styles, extract style features from the sample texts, generate initial feature tags (e.g., "tech-savvy + cyberpunk"), and create a custom style. Furthermore, users can manually adjust the initial feature tags via the interactive interface to obtain personalized feature tags, which are then stored to form a new strategy guide style template for subsequent use in game strategy guide generation. By adding personalized strategy guide style templates, the scalability of game strategy guide generation is improved.

[0069] S12. Based on the user's needs information and the set knowledge base, construct a knowledge graph to obtain key prompts.

[0070] Based on the received user request information and the established knowledge base, a knowledge search is performed to find relevant materials. A knowledge graph is then constructed based on the search results to obtain associated prompts. For example, when the user request information is a text description, word segmentation, part-of-speech tagging, and named entity recognition are performed to extract entities such as game name, game character, game skill / mechanic, and strategy type. For instance, if the user request information is "a strategy guide for character B's shield in Game A, suitable for beginners," word segmentation, part-of-speech tagging, and named entity recognition can yield the following: Game name: Game A; Game character: Game Character B; Game skill: Shield; Strategy type: Team composition. A shield is a temporary defensive measure in a game, protecting a character / unit by absorbing or mitigating damage, usually existing as an energy field or special skill. While the implementation of shields varies across different games, their core function is similar. Shield team composition refers to a strategy in the game that uses specific character or equipment combinations to build a team centered around shield protection, primarily achieving damage reduction, interruption resistance, and buff synergy. Damage reduction mitigates damage from enemy attacks, interruption resistance maintains the stability of character skill casting, and buff synergy triggers damage / healing bonuses related to the shield. When user input is image-based, image recognition technology identifies entities such as the game name, game character, game skill / mechanic, and strategy type. Based on the identified entities, a knowledge graph is constructed using a set knowledge base to obtain key prompts. For example, the key prompts associated with game character B include "game character B," "shield," "team composition," and "equipment 1." "Equipment 1" is associated with the "shield" skill of "game character B," and "equipment 1" enhances the "shield" effect.

[0071] As described above, by using a knowledge base to retrieve and integrate multi-source data, the reliability of data sources for game guide generation is improved, thereby enhancing the accuracy of the generated game guides. Furthermore, it can automatically match corresponding entity data in the knowledge base based on user needs, generating relevant keyword prompts without manual retrieval and matching. This significantly improves the automation and intelligence of guide retrieval and matching, thus increasing the overall efficiency of game guide generation. In addition, the knowledge base supports incremental updates. When adding new characters or new game versions, only the corresponding nodes and relationships need to be added, without restructuring the architecture, thereby improving the scalability and maintainability of game guide generation.

[0072] In one embodiment, a knowledge base needs to be established before constructing the knowledge graph. Game reference materials can be received, including one or more of the following: game standard definitions, game encyclopedia data, cost-based strategies for this game, general game finished product strategies, and restrictive rules. Reference materials can be manually uploaded by users or obtained from network or cloud databases. Game standard definitions are used to interpret various in-game terms; game encyclopedia data is used to explain fixed information such as skills and values; cost-based strategies for this game record strategies already produced for this game; general game finished product strategies record strategies for games other than this game that can be used as references; and restrictive rules record some custom rules. The received game reference materials may exist in various formats, such as text, data, image, and video formats, and the system can support receiving game reference materials in multiple formats. The received game reference materials are parsed to obtain metadata. This metadata includes corresponding game terms, video screenshots, or video files, where game terms include character terms, skill terms, equipment terms, and proprietary terminology. The parsed metadata is annotated to generate the corresponding knowledge base. For example, metadata can be labeled with multiple tags such as game name, game version, character, skill, and value. For instance, if the game reference describes "Character B's E skill shield absorption is linked to maximum health", the corresponding annotation would be: [Character: Character B]'s [Skill: E Skill][Mechanism: Shield Absorption] and [Attribute: Maximum Health][Relationship: Linked]. Deeper semantic annotation is also possible, such as labeling "Skill → Effect" and "Character → Counter Relationship", which are stored in a graph database to obtain the corresponding knowledge base. When constructing the knowledge base, five core entities can be defined: game (name), character, skill, mechanism, and value. Relationships between entities include various types such as belonging, counter, enhancement, and trigger. Based on the aforementioned defined entities and relationship types, triplet data is constructed. For example, triplet data could be: (Character B, belongs to, "Game A" character), (Shield absorption, enhancement, maximum health, +40%), (E skill, trigger, shield generation, 100%). Based on the constructed triplet data, a knowledge base is generated. Subsequent searches within this knowledge base can retrieve corresponding triplet data. As described above, the fusion of multi-source game reference materials enriches the knowledge base, providing abundant reference resources for subsequent game guides generated based on it, thereby improving the richness and accuracy of the generated game guides.

[0073] In one embodiment, knowledge base updates can be processed asynchronously via a message queue. When a new document is added, it is first written to a memory cache. Then, based on the parsing of the new document, the corresponding metadata is obtained, the metadata is annotated, and added to the knowledge base.

[0074] S13. Generate executable strategy generation commands based on key prompts and strategy style templates, and execute the strategy generation commands to generate game strategies of the corresponding style.

[0075] Based on the aforementioned key prompts and the determined strategy guide style template, the key prompts and strategy guide style template are semantically concatenated to form structured prompts. Corresponding executable strategy guide generation instructions are then generated based on these structured prompts. These instructions include sub-instructions for text generation, image retrieval, image-text integration, video retrieval, video compositing, and video-text fusion. Executing these instructions generates game guides in the corresponding style. These game guides can be plain text, image-text combined, or video-text combined. This method automatically generates game guides in the corresponding style using keyword prompts and strategy guide style templates, eliminating the need for manual writing and significantly improving the efficiency of game guide generation.

[0076] As described above, during game guide generation, a guide style template is determined based on the received style selection instruction. Key prompts are obtained by constructing a knowledge graph based on a set knowledge base using the received user requirement information. An executable guide generation instruction is then generated based on the key prompts and the guide style template. Executing the guide generation instruction generates a game guide in the corresponding style. Compared to manually writing game guides, this embodiment can automatically generate game guides in the corresponding style using user requirement information and style selection instructions, improving the automation and intelligence of game guide generation and significantly increasing its efficiency.

[0077] Figure 2 This is a flowchart illustrating another game strategy generation method provided in this application embodiment, see below. Figure 2 The specific methods for generating this game guide include:

[0078] S111: Receive voice information, text descriptions, or images input by the user.

[0079] In the aforementioned S11, when receiving user request information, the user request information can be determined by receiving user input language information, text descriptions, or images. Figure 3 This is a first schematic diagram of a game strategy generation interactive interface provided in an embodiment of this application, referring to... Figure 3The game strategy generation interface 1 displays a voice input control 11, a text input control 12, and an image input control 13. Users can input corresponding information by triggering the corresponding control. For example, triggering the voice input control 11 allows for voice input, triggering the text input control 12 allows for text description input, and triggering the image input control 13 allows for image input. This multi-faceted approach to inputting user needs enhances the diversity of interaction, allowing users to input their requirements in different ways and thus improving the user experience.

[0080] S112. Perform text parsing processing on the received voice information to obtain the corresponding user requirement information.

[0081] When a user inputs their needs via voice, the system receives the voice input and performs text parsing to obtain the corresponding user request information. For example, the received voice information undergoes format standardization and noise reduction to obtain standardized voice information. The standardized voice information is then segmented into frames of a preset size (e.g., 20-30ms) using a Hamming window to reduce spectral leakage and prepare for subsequent feature extraction. Acoustic feature extraction and text conversion are performed on the aforementioned voice frames to obtain the corresponding text description. The text description is then semantically adjusted and context-completed to obtain the corresponding user request information. For example, the obtained user request information might be "A guide to character B's shield team composition in Game A, suitable for beginners." As described above, voice input significantly improves input efficiency compared to text input, thereby increasing the efficiency of generating the corresponding game guide.

[0082] S113. Perform semantic analysis on the received text description to obtain the corresponding user requirement information.

[0083] When a user inputs their needs via text, the system receives the text description and performs semantic analysis to identify entities and key information within it. Entities include the game name, character name, mechanic terminology, version number, strategy direction, and target audience. Key information includes relationships between entities. For example, if the input is "Character B's Abyss Team Composition Guide in Version 3.8," the corresponding relationships are: "Character B - Belongs to - Team Composition, Version 3.8 - Limited - Abyss." Based on the entities and relationships within the text description, semantic expansion and context completion are performed to obtain the corresponding user needs information. This process of text parsing and processing the received text description to obtain corresponding user needs information enables automatic expansion and semantic understanding of text input, improving the accuracy of identifying user needs and consequently enhancing the accuracy of subsequent game guides generated based on those needs.

[0084] S114. Perform intelligent recognition processing on the received image to obtain the corresponding user demand information.

[0085] When a user inputs their request information via an image, the system receives the input image, which can be a screenshot of the corresponding game interface. The received image undergoes intelligent recognition processing to identify game elements and text information. Game elements include character avatars, skill icons, and equipment interfaces. Semantic association analysis is performed based on the identified game elements and text information. The results are then integrated to output the corresponding user request information. For example, if the identified game elements are character B, skill a, and equipment 1, and the corresponding identified text information is "Character B," "Shield Enhancement," and "20%," then integrating the game elements and text information, the output user request information is "Shield Team Composition Guide for Character B in Game A." This method of determining user request information through image input is faster than text descriptions, thus improving the efficiency of identifying user request information and consequently increasing the efficiency of generating corresponding game guides based on that information.

[0086] In one embodiment, when determining user demand information based on the received voice information, text description, or image, if the user demand information cannot be accurately determined, follow-up questions can be asked to remind the user to supplement the missing information. Subsequently, complete user demand information can be generated based on the answers to the follow-up questions and the aforementioned voice information, text description, or image.

[0087] In one embodiment, the user can also simultaneously input voice information, text descriptions, and images. The system receives the user's input voice information, text descriptions, and images; performs text parsing on the received voice information to obtain first requirement information; performs semantic analysis on the received text descriptions to obtain second requirement information; and performs intelligent recognition on the received images to obtain third requirement information. The first, second, and third requirement information are then integrated to obtain the corresponding user requirement information.

[0088] The above provides three input interaction methods: voice information, text description, and images. This allows users to input their needs using multiple interaction methods, improving interaction flexibility and thus enhancing the user experience.

[0089] Figure 4 This is a flowchart illustrating another game strategy generation method provided in this application embodiment, see below. Figure 4 The methods for generating this game guide include:

[0090] S121. Determine the search index based on user needs information.

[0091] After receiving the user request information in S11, a search index is determined based on the user request information. For example, the search index can be entities from the user request information, such as game name, character name, skill name, etc. For instance, if the user request information is "a guide to team composition for character B's shield in Game A, suitable for beginners," then the corresponding index information includes Game A, character B, shield, and team composition, etc.

[0092] S122. Based on the set knowledge base, retrieve metadata according to the retrieval index to obtain the target triplet data.

[0093] The target triplet data is determined by retrieving metadata from the knowledge base based on the retrieval index. Since the corresponding triplet data has already been semantically annotated in the knowledge base, the target triplet data can be retrieved by retrieving metadata from the knowledge base using the retrieval index. For example, if the retrieval index is "Game A and Character B", then the target triplet data for "Game A and Character B" can be retrieved from the knowledge base. For instance, the retrieved metadata is parsed to extract the entities and relationships between them, forming the target triplet data based on these relationships. For example, if the retrieved metadata is "In version 3.8 of Game A, Character B's shield mechanism is related to Equipment 1", then the extracted target triplet data would be (Character B, uses, Equipment 1), (Character B, belongs to, Game A), and (Shield mechanism, is, Character B's skill), etc. Since the retrieved target triplet data may come from different data sources, the target triplet data can be fused to combine target triplet data from different data sources.

[0094] S123. Construct a knowledge graph based on the target triplet data, and extract key prompts from the constructed knowledge graph.

[0095] The knowledge graph is stored and constructed based on a set graph database. This involves importing target triplet data into the graph database, creating nodes and edges, and generating the final knowledge graph. The nodes in the constructed knowledge graph are ranked by importance, and the top-ranked nodes are selected as candidate keywords. For example, in the knowledge graph of character B in game A, nodes such as "Character B," "Equipment 1," and "Shield Mechanism" rank highly. The candidate keywords are then filtered and adjusted based on the semantics and context of the user's needs. For example, if the user's needs emphasize beginner-friendly guides, keywords such as "beginner" and "basic" are added. The final extracted keyword is: "Beginner's Abyss Team Building Guide for Character B in Version 3.8 of Game A, Detailed Explanation of the Shield Mechanism of Equipment 1."

[0096] As described above, by determining the retrieval index based on user needs information, and then using this precise index to efficiently retrieve metadata based on a set knowledge base, data retrieval time is significantly shortened, and metadata retrieval efficiency is improved, thereby increasing the efficiency of game guide generation. Keyword prompts extracted based on knowledge graphs and user needs information accurately reflect the core content of user needs, enabling subsequently generated game guides to better meet user requirements, improving the accuracy of game guide generation, and ultimately enhancing the user experience.

[0097] Figure 5 This is a flowchart illustrating another game strategy generation method provided in this application embodiment, see below. Figure 5 The specific methods for generating this game guide include:

[0098] S131. Determine the text generation sub-instruction, image acquisition sub-instruction, and image-text integration sub-instruction in the strategy generation command.

[0099] The type of game guide generated can be determined based on user needs. It can be plain text, a combination of text and images, or a combination of video and text. When the user requests a combination of text and images, the executable guide generation instructions mentioned above are defined as follows: text generation sub-instruction, image acquisition sub-instruction, and text-image integration sub-instruction. Specifically, the text generation sub-instruction generates the game guide text, the image acquisition sub-instruction retrieves corresponding game interface screenshots, and the text-image integration sub-instruction merges the text and images to generate the final text-image game guide.

[0100] In one embodiment, when the game guide requested by the user is plain text, the aforementioned guide generation instruction is executed, and the corresponding style of game guide text is generated by combining the guide style template (text type template) with the key prompts generated by the knowledge graph.

[0101] S132. Execute the text generation sub-instruction to generate game strategy text in the corresponding style.

[0102] The text generation sub-instruction is executed, combining the strategy style template (text-type template) with key prompts generated from the knowledge graph to generate game strategy text in the corresponding style. For example, dynamic parameters are determined in the key prompts, where dynamic parameters refer to replaceable variable parts (i.e., template variables) in the key prompts, such as game name, game character, and version. For example, the "humorous style" template is: "Family members, who understands this! {{character}}'s {{mechanism}} is too {{adjective}}! Newbies are advised to just use {{verb}}, it's practically {{noun}}~". Here, character, mechanism, adjective, verb, and noun are template variables, establishing a mapping relationship between template variables and dynamic parameters. For example, character corresponds to "character B", mechanism corresponds to "shield", adjective corresponds to "thick", verb corresponds to "open E", and noun corresponds to "escape artifact". The key prompts corresponding to the dynamic parameters are filled into the strategy style template to generate an executable text generation sub-instruction. For example, the variable placeholders in the strategy style template can be parsed, and the value corresponding to each variable placeholder, such as "Character B," can be obtained from the aforementioned mapping relationship. The string corresponding to the strategy style template is executed to generate a complete executable text generation sub-instruction. This text generation sub-instruction is then executed to generate and output the corresponding style of game strategy text. Alternatively, the text generation sub-instruction can be converted into an executable API call, which calls the corresponding text generation model to generate the corresponding game strategy text. For example, the generated game strategy text might be: "Fellow players, who understands! Character B's shield is too thick! Newbies are advised to directly activate E; it's practically an escape tool~."

[0103] As mentioned above, automatically generating game strategy text in the corresponding style through text generation sub-commands greatly improves the efficiency of game strategy text generation compared to manual writing.

[0104] S133. Execute the image acquisition sub-instruction to retrieve the corresponding game screenshot from the set knowledge base based on the game strategy text, and obtain the target image.

[0105] Executing the aforementioned image retrieval sub-instruction, the generated game strategy text undergoes keyword extraction and parsing to obtain the corresponding target keywords. For example, semantic analysis and keyword recognition are performed on the game strategy text to identify key entities, distinguish categories such as characters, equipment, and mechanics, and extract attribute words such as "shield enhancement" and "health bonus." Since each game screenshot in the set knowledge base has been annotated with metadata, the corresponding target image can be retrieved based on the metadata annotation using the aforementioned target keywords. For example, images containing all target keywords are retrieved first; then, a semantic matching model is used to calculate the cross-modal similarity between the game strategy text and the image, obtaining images with a similarity greater than a preset threshold; finally, an image recognition model is used to detect characters and equipment in the image, filtering out irrelevant content (i.e., screenshots) to obtain the final target image.

[0106] As mentioned above, the automatic acquisition of game screenshots related to game strategy text through image acquisition sub-commands greatly improves the efficiency of image acquisition in text-based game strategy guides compared to manual retrieval or manual screenshotting, thereby improving the final efficiency of game strategy guide generation.

[0107] S134. Execute the image and text integration sub-instruction to perform image and text fusion processing on the game strategy text and the target image to obtain the first target game strategy.

[0108] Execute the aforementioned sub-instruction for integrating text and images to obtain text keywords from the game guide text. Calculate the semantic similarity between the text keywords and images to determine the strength of the text-image association. For example, the text keyword "shield effect of Equipment 1" and the image feature "Equipment 1 interface and shield generation animation" have a calculated semantic similarity of 0.92. Assuming a preset threshold of 0.7, the calculated semantic similarity is greater than the preset threshold, indicating a strong text-image association. Establish an association mapping relationship between the image and text keywords with the strongest text-image association. Generate a corresponding image anchor for each text keyword in the game guide text. For example, clicking "Equipment 1 effect" will automatically jump to the image displaying that attribute. Perform text-image fusion processing on the game guide text and target images, and fuse the target images with corresponding text keywords that have an association mapping relationship. Text-image fusion can be performed according to preset layout templates, such as tutorial templates, display templates, and data templates. The tutorial template displays images on the left and game guide text on the right; the presentation template displays images at the top and game guide text at the bottom; the data template displays images and game guide text synchronously in a table. After the image and text fusion process is completed, the first target game guide combining images and text is output.

[0109] As mentioned above, the automatic generation of corresponding text and image-based game guides using the guide generation command significantly improves the efficiency of game guide generation compared to manual writing. Furthermore, presenting game guides using a combination of text and images allows users to better understand and remember key information, improving information delivery efficiency and ultimately enhancing the user experience.

[0110] Figure 6 This is a flowchart illustrating another game strategy generation method provided in this application embodiment, see below. Figure 6 The specific methods for generating this game guide include:

[0111] S135, Determine the text generation sub-instruction, video acquisition sub-instruction, video compositing sub-instruction, and video-text fusion sub-instruction in the strategy generation instruction.

[0112] The type of game guide generated can be determined based on user needs. It can be plain text, a combination of text and images, or a combination of video and text. When the user requests a video-text combination game guide, the following sub-instructions are defined in the aforementioned guide generation command: text generation, video acquisition, video compositing, and video-text fusion. Specifically, the text generation sub-instruction generates the game guide text, the video acquisition sub-instruction acquires the corresponding game video, the video compositing sub-instruction performs cutting and compositing on the acquired game video, and the video-text fusion sub-instruction merges the game guide text and the composited video to generate the final video-text combination game guide.

[0113] S136. Execute the text generation sub-command to generate game strategy text in the corresponding style.

[0114] The text generation sub-instruction is executed, combining the strategy style template (text-type template) with key prompts generated from the knowledge graph to generate game strategy text in the corresponding style. For example, dynamic parameters are determined in the key prompts, where dynamic parameters refer to replaceable variable parts (i.e., template variables) in the key prompts, such as game name, game character, and version. For example, the "humorous style" template is: "Family members, who understands this! {{character}}'s {{mechanism}} is too {{adjective}}! Newbies are advised to just use {{verb}}, it's practically {{noun}}~". Here, character, mechanism, adjective, verb, and noun are template variables, establishing a mapping relationship between template variables and dynamic parameters. For example, character corresponds to "character B", mechanism corresponds to "shield", adjective corresponds to "thick", verb corresponds to "open E", and noun corresponds to "escape artifact". The key prompts corresponding to the dynamic parameters are filled into the strategy style template to generate an executable text generation sub-instruction. For example, the variable placeholders in the strategy style template can be parsed, and the value corresponding to each variable placeholder, such as "Character B," can be obtained from the aforementioned mapping relationship. The string corresponding to the strategy style template is executed to generate a complete executable text generation sub-instruction. This text generation sub-instruction is then executed to generate and output the corresponding style of game strategy text. Alternatively, the text generation sub-instruction can be converted into an executable API call, which calls the corresponding text generation model to generate the corresponding game strategy text. For example, the generated game strategy text might be: "Fellow players, who understands! Character B's shield is too thick! Newbies are advised to directly activate E; it's practically an escape tool~."

[0115] As mentioned above, automatically generating game strategy text in the corresponding style through text generation sub-commands greatly improves the efficiency of game strategy text generation compared to manual writing.

[0116] S137. Execute the video acquisition sub-instruction to retrieve the corresponding target game video from the set knowledge base according to the user's requirements information, and edit the target game video according to the key prompts to obtain the target video segment.

[0117] The process executes a video acquisition sub-instruction to extract entities from user request information, including game name, character, mechanics, and version. Based on these entities, a game video search is performed from a pre-defined knowledge base, retrieving videos that match the entities. Then, the cosine similarity between the entity and video features is calculated, identifying videos with a cosine similarity greater than a preset threshold. Next, a visual detection model is used to detect characters and equipment in keyframes of the video, identifying videos where the characters and equipment in the keyframes match the aforementioned entities, thus obtaining the final target game video. The process matches the timestamps in the metadata of the target game video with the key prompts determined in S12. For example, if the key prompt is "the shield generation timing of character B's equipment 1," the corresponding timestamp of the target game video is "equipment 1 effect_00:01:10-00:01:30." An editing script is generated based on the aforementioned key prompts and corresponding timestamps, and the target game video is edited using this script to obtain the corresponding target video clip.

[0118] The above-mentioned method of automatically acquiring target game videos and automatically editing them into target video segments that match the key prompts corresponding to the user's needs through video acquisition sub-commands greatly improves the efficiency of video acquisition and video editing compared to manual video acquisition and editing methods.

[0119] S138. Execute the video compositing sub-instruction to dynamically composite the target video clips based on the game strategy text to obtain the initial strategy video.

[0120] The video compositing sub-instruction executes semantic segmentation of the game guide text, resulting in segmented text. Timelines are then assigned to these segmented texts based on their textual logic. Dynamic parameters (such as character, skill, and equipment information) are extracted from the game guide text and associated with target video segments. Based on the assigned timelines, the corresponding target video segments are seamlessly stitched together to obtain the initial guide video. For example, corresponding transition effects, such as hard cuts, fade-in, and fade-out, can be added to each target video segment.

[0121] The above-mentioned method of automatically synthesizing target video clips into an initial strategy video that matches the game strategy text through video synthesis sub-instructions greatly improves the efficiency of strategy video synthesis compared to manual synthesis methods.

[0122] S139. Execute the video text fusion sub-instruction to embed the game strategy text into the initial strategy video to obtain the second target game strategy.

[0123] The video-text fusion sub-instruction is executed to semantically segment the game guide text, resulting in segmented text. The semantic similarity between the text keywords in the segmented text and the video frames in the initial guide video is calculated. Video frames with a semantic similarity greater than a preset threshold are identified as associated video frames with the corresponding text keywords, thus determining the timing of keyword display. Based on the display timing, the corresponding segmented text is embedded into the initial guide video to form video subtitles, resulting in the second target game guide of the video-text fusion type.

[0124] The aforementioned automatic video-text combination of walkthrough generation commands significantly improves the efficiency of game walkthrough generation compared to manually editing video walkthroughs. Furthermore, presenting game walkthroughs through a combination of video and text allows users to better understand and remember key information, improving information delivery efficiency and ultimately enhancing the user experience.

[0125] Figure 7 This is a flowchart illustrating another game strategy generation method provided in this application embodiment, see below. Figure 7 The specific methods for generating this game guide include:

[0126] S21. Display the game guide and its corresponding style index and professionalism index on the interactive interface.

[0127] After generating the corresponding style of game guide in S13, the game guide, along with its corresponding style index and professionalism index, can be displayed on the interactive interface. For example, language features are extracted from the game guide to extract style-related features. For instance, if the style is "humorous," the corresponding style-related features include the density of emoticons and the proportion of trending internet terms; if the style is "academic," the corresponding style-related features include the density of technical terms and the proportion of long sentences. Based on the extracted style-related features and the set calculation model, the style index is calculated and output, with a value ranging from 0% to 100%. The numerical values ​​in the game guide are factually verified based on the set knowledge base to determine the numerical accuracy coefficient; the hierarchy of the game guide is analyzed to determine the hierarchy coefficient; and the technical terms in the game guide are analyzed to determine the terminology consistency coefficient. The professionalism index is then calculated based on the numerical accuracy coefficient, hierarchy coefficient, and terminology consistency coefficient. For example, the professionalism index = numerical accuracy coefficient × weight (e.g., 40%) + hierarchy coefficient × weight (e.g., 30%) + terminology consistency coefficient × weight (e.g., 30%).

[0128] Figure 8 This is a second schematic diagram of a game strategy generation interactive interface provided in an embodiment of this application, referring to... Figure 8After obtaining the style index and professional index, the interactive interface displays the game guide along with its corresponding style index and professional index. For example, ... Figure 8 As shown, the style index of this game guide is 75%, and the professionalism index is 50%.

[0129] As mentioned above, by displaying the style index and professionalism index of the generated game guide in the interactive interface, users can easily see whether the style index and professionalism index of the game guide meet their needs. If not, they can adjust the style index and professionalism index of the game guide according to their own needs, thereby improving the user experience.

[0130] S22. Receive adjustment instructions based on style index and professionalism index, and obtain target style index and target professionalism index according to the adjustment instructions.

[0131] After the game guide and its corresponding style and professionalism indices are displayed on the interactive interface, if the displayed style and professionalism indices do not meet the user's needs, the user can adjust the style and professionalism indices accordingly based on the interactive interface. (Refer to...) Figure 8 Users can drag the progress bars corresponding to the style index and professionalism index to input corresponding adjustment commands. The system receives these adjustment commands based on the style index and professionalism index, and determines the target style index and target professionalism index accordingly. For example, if the original style index is 75%, dragging the style index progress bar to 85% will result in a target style index of 85%; if the original professionalism index is 50%, dragging the professionalism index progress bar to 90% will result in a target professionalism index of 90%.

[0132] As described above, by adjusting the style and professionalism indices of the generated game guides using adjustment commands, the adjusted target style and professionalism indices can meet user needs, thereby improving the user experience. Furthermore, through this simple adjustment of the style and professionalism indices, new game guides can be automatically generated based on the target style and professionalism indices, eliminating the need for manual rewriting or adjustment. This significantly improves the efficiency of game guide adjustments and further enhances the user experience.

[0133] S23. Based on the target style index and target professionalism index, and using user needs information and style selection instructions, generate a new game guide.

[0134] Comparing the target style index with the original style index, if the target style index is greater than the original style index, the strategy is to increase slang, emojis, and short sentence rhythms; if the target style index is less than the original style index, the strategy is to reduce slang, emojis, and short sentence rhythms. Comparing the target professionalism index with the original professionalism index, if the target professionalism index is greater than the original professionalism index, the strategy is to add mechanism decomposition and numerical formulas; if the target professionalism index is less than the original professionalism index, the strategy is to reduce mechanism decomposition and numerical formulas. For example, the target style index can be decomposed into executable language feature parameters based on the determined strategy, such as: 20 emojis per 100 characters, 30% of trending internet words, and 60% irony intensity. The target professionalism index can be decomposed into content depth parameters based on the determined strategy, such as: 25% professional terminology, 3 formulas, and version comparison analysis. Based on the decomposed language feature parameters and content depth parameters, key prompts are reconstructed based on user needs information. For example, the original keyword (style index 75%+, professionalism index 50): "Character B's shield strategy"; the new keyword: "Using internet slang like 'XXX' (style (humor) index 90%+), combined with a comparison of shield formulas in versions 3.8 / 4.0 (professionalism index 85%+), providing a detailed analysis of character B's shield threshold calculation (including 3 formulas)." Based on the new keyword, a new game strategy with the corresponding style is regenerated, ensuring that the newly generated game strategy meets both the target style index and the target professionalism index.

[0135] As described above, the generation of new game strategies driven by style and professionalism indices significantly improves the efficiency of game strategy adjustments compared to manual writing and modification. Furthermore, the style and professionalism indices can be adjusted according to the user's individual needs, ensuring that the new game guides generated based on the adjusted target style and professionalism indices meet the user's personalized requirements, greatly enhancing the user experience.

[0136] Based on the above embodiments, Figure 9 This is a schematic diagram of a game strategy generation device provided in an embodiment of this application. (Reference) Figure 9 The game strategy generation device provided in this embodiment specifically includes: an information receiving module 21, a template determining module 22, a prompt word determining module 23, and a strategy generation module 24.

[0137] Among them, the information receiving module 21 is used to receive user demand information and style selection instructions;

[0138] Template determination module 22 is used to determine the strategy style template based on the style selection instruction. User requirement information includes one or more of the following: game name, game version, game character, strategy direction, and target audience.

[0139] The prompt word determination module 23 is used to construct a knowledge graph based on the set knowledge base according to user needs information to obtain key prompt words;

[0140] The strategy generation module 24 is used to generate executable strategy generation instructions based on key prompts and strategy style templates, and to generate game strategies of the corresponding style by executing the strategy generation instructions.

[0141] In one embodiment, the game strategy generation device further includes: a reference material receiving module, a reference material parsing module, and a knowledge base generation module;

[0142] The reference material receiving module is used to receive game reference materials, which include one or more of the following: game standard definitions, game encyclopedia data, cost guides for this game, guides for finished general games, and restriction rules;

[0143] The reference data parsing module is used to parse game reference data to obtain metadata;

[0144] The knowledge base generation module is used to annotate metadata and generate the corresponding knowledge base.

[0145] In one embodiment, the prompt word determination module 23 includes: an index determination submodule, a retrieval submodule, a graph construction submodule, and a key prompt word determination submodule;

[0146] The index determination submodule is used to determine the retrieval index based on user requirements.

[0147] The retrieval submodule is used to retrieve metadata based on the set knowledge base according to the retrieval index to obtain the target triplet data;

[0148] The graph construction submodule is used to construct a knowledge graph based on the target triple data.

[0149] The Keyword Determination Submodule is used to extract key words from the constructed knowledge graph.

[0150] In one embodiment, the strategy generation module 24 includes: a sub-instruction determination sub-module, a text generation sub-module, an image acquisition sub-module, and an image-text fusion sub-module;

[0151] The sub-instruction determination sub-module is used to determine the text generation sub-instruction, image acquisition sub-instruction, and image-text integration sub-instruction in the strategy generation instruction;

[0152] The text generation submodule is used to execute text generation sub-instructions to generate game strategy text in the corresponding style;

[0153] The image acquisition submodule is used to execute image acquisition sub-instructions, which retrieve the corresponding game screenshots from the set knowledge base based on the game strategy text, and obtain the target image;

[0154] The image and text fusion submodule is used to execute the image and text integration sub-instruction, which merges the game strategy text and the target image to obtain the first target game strategy.

[0155] In one embodiment, the strategy generation module 24 further includes: a video acquisition submodule, a video editing submodule, a video compositing submodule, and a subtitle fusion submodule;

[0156] The sub-instruction determination sub-module is also used to determine the text generation sub-instruction, video acquisition sub-instruction, video synthesis sub-instruction, and video-text fusion sub-instruction in the strategy generation instruction;

[0157] The text generation submodule is also used to execute text generation sub-instructions to generate game strategy text in the corresponding style;

[0158] The video acquisition submodule is used to execute video acquisition sub-instructions and retrieve the corresponding target game video from the set knowledge base according to the user's requirements.

[0159] The video editing submodule is used to edit the target game video based on key prompts to obtain target video clips;

[0160] The video compositing submodule is used to execute video compositing sub-instructions, dynamically compositing the target video clips based on the game strategy text to obtain the initial strategy video;

[0161] The subtitle fusion submodule is used to execute video text fusion sub-instructions, embedding game strategy text into the initial strategy video to obtain the second target game strategy.

[0162] In one embodiment, the game strategy generation device further includes: a display module, an adjustment module, and a strategy regeneration module;

[0163] The display module is used to display game guides and their corresponding style and professionalism indices on the interactive interface.

[0164] The adjustment module is used to receive adjustment instructions based on the style index and the professionalism index, and to obtain the target style index and the target professionalism index according to the adjustment instructions;

[0165] The strategy regeneration module is used to regenerate new game strategies based on user needs information and style selection instructions, according to the target style index and target professionalism index.

[0166] In one embodiment, the information receiving module 21 includes: an information receiving submodule, a voice parsing submodule, a text parsing submodule, and an image recognition submodule;

[0167] The information receiving submodule is used to receive user input, such as voice messages, text descriptions, or images.

[0168] The speech parsing submodule is used to perform text parsing processing on the received speech information to obtain the corresponding user requirement information;

[0169] The text parsing submodule is used to perform semantic analysis on the received text descriptions to obtain the corresponding user requirement information;

[0170] The image recognition submodule is used to intelligently recognize and process received images to obtain corresponding user requirement information.

[0171] The game strategy generation device provided in this application embodiment can be used to execute the game strategy generation method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0172] This application provides a game strategy generation device, referring to... Figure 10 The game strategy generation device includes: a processor 31, a memory 32, a communication module 33, an input device 34, and an output device 35. The game strategy generation device may have one or more processors, and the game strategy generation device may have one or more memories. The processor, memory, communication module, input device, and output device of the game strategy generation device can be connected via a bus or other means.

[0173] The memory 32, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as program instructions / modules corresponding to the game strategy generation method in any embodiment of this application (e.g., the information receiving module, template determining module, prompt word determining module, and strategy generation module in the game strategy generation device). The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on the use of the device, etc. Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0174] The communication module 33 is used for data transmission.

[0175] The processor 31 executes various functional applications and data processing of the device by running software programs, instructions and modules stored in the memory, thereby realizing the above-mentioned game strategy generation method.

[0176] Input device 34 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the device. Output device 35 may include display devices such as a display screen.

[0177] The game strategy generation device provided above can be used to execute the game strategy generation method provided in the above embodiments, and has corresponding functions and beneficial effects.

[0178] This application embodiment also provides a storage medium for storing computer-executable instructions. When executed by a computer processor, the computer-executable instructions are used to execute a game guide generation method. The game guide generation method includes: receiving user demand information and a style selection instruction; determining a guide style template based on the style selection instruction; wherein the user demand information includes one or more of the following: game name, game version, game character, guide direction, and target audience; constructing a knowledge graph based on a set knowledge base according to the user demand information to obtain key prompts; generating an executable guide generation instruction based on the key prompts and the guide style template; and executing the guide generation instruction to generate a game guide of the corresponding style.

[0179] Storage medium – any type of memory device or storage device. The term “storage medium” is intended to include: mounting media, such as CD-ROM, floppy disk, or magnetic tape devices; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. Storage medium may also include other types of memory or combinations thereof. Furthermore, storage medium may reside in a first computer system in which the program is executed, or it may reside in a different second computer system connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term “storage medium” can include two or more storage media residing in different locations (e.g., in different computer systems connected via a network). Storage medium may store program instructions (e.g., specifically implemented as a computer program) executable by one or more processors.

[0180] Of course, the computer-executable instructions stored in the storage medium provided in the embodiments of this application are not limited to the game strategy generation method described above, but can also execute related operations in the game strategy generation method provided in any embodiment of this application.

[0181] The game strategy generation device, storage medium, and game strategy generation equipment provided in the above embodiments can execute the game strategy generation method provided in any embodiment of this application. For technical details not described in detail in the above embodiments, please refer to the game strategy generation method provided in any embodiment of this application.

[0182] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. A game guidebook generation method characterized by comprising: The method comprises the following steps: receiving user demand information and style selection instructions, determining a guide style template according to the style selection instructions, the user demand information including one or more of game name, game version, game character, guide direction and target audience; constructing a knowledge graph based on the set knowledge base according to the user demand information, obtaining key prompt words, including: determining a retrieval index according to the user demand information, retrieving metadata based on the set knowledge base according to the retrieval index, obtaining target triple data, constructing a knowledge graph according to the target triple data, and extracting key prompt words from the constructed knowledge graph; generating executable guide generation instructions according to the key prompt words and the guide style template, including: semantic splicing the key prompt words and the guide style template to form structured prompt words, and generating corresponding executable guide generation instructions according to the structured prompt words; executing the guide generation instructions to generate game guides of corresponding styles, including: determining text generation sub-instructions, video acquisition sub-instructions, video synthesis sub-instructions and video and text fusion sub-instructions in the guide generation instructions, executing the text generation sub-instructions to generate game guide texts of corresponding styles, executing the video acquisition sub-instructions to acquire corresponding target game videos from the set knowledge base according to the user demand information, and editing the target game videos according to the key prompt words to obtain target video clips, executing the video synthesis sub-instructions to dynamically synthesize and process the target video clips according to the game guide texts to obtain initial guide videos, and executing the video and text fusion sub-instructions to embed the game guide texts into the initial guide videos to obtain second target game guides.

2. The method of claim 1, wherein, Before the step of constructing a knowledge graph based on the set knowledge base according to the user demand information, the method further comprises the following steps: receiving game reference materials, the reference materials including one or more of game standard definition, game encyclopedia data, game cost guide, general game product guide and restriction rules; analyzing the game reference materials to obtain metadata, and labeling the metadata to generate a corresponding knowledge base.

3. The method of claim 1, wherein, The step of executing the guide generation instructions to generate game guides of corresponding styles further comprises the following steps: determining text generation sub-instructions, picture acquisition sub-instructions and picture and text integration sub-instructions in the guide generation instructions; executing the text generation sub-instructions to generate game guide texts of corresponding styles; executing the picture acquisition sub-instructions to acquire corresponding game screenshots from the set knowledge base according to the game guide texts to obtain target pictures; executing the picture and text integration sub-instructions to perform picture and text fusion processing on the game guide texts and the target pictures to obtain first target game guides.

4. The method of claim 1, wherein, After the step of executing the guide generation instructions to generate game guides of corresponding styles, the method further comprises the following steps: displaying the game guides, style indexes and professional degree indexes corresponding to the game guides in an interactive interface; receive an adjustment instruction based on the style index and the professionalism index, and obtain a target style index and a target professionalism index according to the adjustment instruction; regenerate a new game guide based on user demand information and a style selection instruction according to the target style index and the target professionalism index.

5. The method of claim 1, wherein, The receiving user demand information includes: receiving voice information, text description or picture input by a user; performing text analysis processing on the received voice information to obtain corresponding user demand information; or, performing semantic analysis processing on the received text description to obtain corresponding user demand information; or, performing intelligent recognition processing on the received picture to obtain corresponding user demand information.

6. A game strategy generating apparatus characterized by comprising: It includes: an information receiving module configured to receive user demand information and a style selection instruction; a template determining module configured to determine a guide style template according to the style selection instruction, and the user demand information including one or more of a game name, a game version, a game character, a guide direction and a target audience; a prompt word determining module configured to determine a search index according to the user demand information, perform metadata retrieval based on a set knowledge base according to the search index to obtain target triple data, construct a knowledge graph according to the target triple data, and extract key prompt words from the constructed knowledge graph; a guide generating module configured to generate executable guide generation instructions according to the key prompt words and the guide style template, wherein the executable guide generation instructions include: performing semantic splicing on the key prompt words and the guide style template to form structured prompt words, generating corresponding executable guide generation instructions according to the structured prompt words, determining text generation sub-instructions, video acquisition sub-instructions, video synthesis sub-instructions and video-text fusion sub-instructions in the guide generation instructions, executing the text generation sub-instructions to generate game guide texts in a corresponding style, executing the video acquisition sub-instructions to acquire corresponding target game videos from a set knowledge base according to the user demand information, and performing editing on the target game videos according to the key prompt words to obtain target video clips, executing the video synthesis sub-instructions to perform dynamic synthesis processing on the target video clips according to the game guide texts to obtain initial guide videos, and executing the video-text fusion sub-instructions to embed the game guide texts into the initial guide videos to obtain second target game guides.

7. A game strategy generating device characterized by comprising: It includes: a memory and one or more processors; the memory is configured to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-5.

8. A storage medium storing computer-executable instructions, wherein: The computer executable instructions, when executed by the processor, are used to perform the method of any one of claims 1-5. The computer executable instructions, when executed by the processor, are used to perform the method of any one of claims 1-5.

Images